{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:KTHALUMTW5NU7ZBM3VFN7MWMRB","short_pith_number":"pith:KTHALUMT","schema_version":"1.0","canonical_sha256":"54ce05d193b75b4fe42cdd4adfb2cc885cf03d006ce7da4dd8a2432876aeedbd","source":{"kind":"arxiv","id":"2112.08352","version":2},"attestation_state":"computed","paper":{"title":"Textless Speech-to-Speech Translation on Real Data","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG","eess.AS"],"primary_cat":"cs.CL","authors_text":"Ann Lee, Changhan Wang, Holger Schwenk, Hongyu Gong, Jiatao Gu, Juan Pino, Paul-Ambroise Duquenne, Peng-Jen Chen, Sravya Popuri, Wei-Ning Hsu, Yossi Adi","submitted_at":"2021-12-15T18:56:35Z","abstract_excerpt":"We present a textless speech-to-speech translation (S2ST) system that can translate speech from one language into another language and can be built without the need of any text data. Different from existing work in the literature, we tackle the challenge in modeling multi-speaker target speech and train the systems with real-world S2ST data. The key to our approach is a self-supervised unit-based speech normalization technique, which finetunes a pre-trained speech encoder with paired audios from multiple speakers and a single reference speaker to reduce the variations due to accents, while pre"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2112.08352","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-12-15T18:56:35Z","cross_cats_sorted":["cs.AI","cs.LG","eess.AS"],"title_canon_sha256":"978b52948c00de9d4036a78519c849784aaff492d4b49a409b8df6731e91ef5a","abstract_canon_sha256":"e4ffd974eececba1938195fd4f5a47d1f0610e12fe330e2aab9087edc19cb85c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:20:31.742994Z","signature_b64":"gbIQGM3dIYyrGVnWS1Lf7c3kRhwOwRlqwH1B9S8lDfQSXSVDLuLpS+BmKXoryq6IC3TfpmrJJPeMvNvtn5ySDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"54ce05d193b75b4fe42cdd4adfb2cc885cf03d006ce7da4dd8a2432876aeedbd","last_reissued_at":"2026-07-05T04:20:31.742515Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:20:31.742515Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Textless Speech-to-Speech Translation on Real Data","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG","eess.AS"],"primary_cat":"cs.CL","authors_text":"Ann Lee, Changhan Wang, Holger Schwenk, Hongyu Gong, Jiatao Gu, Juan Pino, Paul-Ambroise Duquenne, Peng-Jen Chen, Sravya Popuri, Wei-Ning Hsu, Yossi Adi","submitted_at":"2021-12-15T18:56:35Z","abstract_excerpt":"We present a textless speech-to-speech translation (S2ST) system that can translate speech from one language into another language and can be built without the need of any text data. Different from existing work in the literature, we tackle the challenge in modeling multi-speaker target speech and train the systems with real-world S2ST data. The key to our approach is a self-supervised unit-based speech normalization technique, which finetunes a pre-trained speech encoder with paired audios from multiple speakers and a single reference speaker to reduce the variations due to accents, while pre"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2112.08352","kind":"arxiv","version":2},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2112.08352/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2112.08352","created_at":"2026-07-05T04:20:31.742576+00:00"},{"alias_kind":"arxiv_version","alias_value":"2112.08352v2","created_at":"2026-07-05T04:20:31.742576+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2112.08352","created_at":"2026-07-05T04:20:31.742576+00:00"},{"alias_kind":"pith_short_12","alias_value":"KTHALUMTW5NU","created_at":"2026-07-05T04:20:31.742576+00:00"},{"alias_kind":"pith_short_16","alias_value":"KTHALUMTW5NU7ZBM","created_at":"2026-07-05T04:20:31.742576+00:00"},{"alias_kind":"pith_short_8","alias_value":"KTHALUMT","created_at":"2026-07-05T04:20:31.742576+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2306.12925","citing_title":"AudioPaLM: A Large Language Model That Can Speak and Listen","ref_index":29,"is_internal_anchor":false},{"citing_arxiv_id":"2507.16632","citing_title":"Step-Audio 2 Technical Report","ref_index":45,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/KTHALUMTW5NU7ZBM3VFN7MWMRB","json":"https://pith.science/pith/KTHALUMTW5NU7ZBM3VFN7MWMRB.json","graph_json":"https://pith.science/api/pith-number/KTHALUMTW5NU7ZBM3VFN7MWMRB/graph.json","events_json":"https://pith.science/api/pith-number/KTHALUMTW5NU7ZBM3VFN7MWMRB/events.json","paper":"https://pith.science/paper/KTHALUMT"},"agent_actions":{"view_html":"https://pith.science/pith/KTHALUMTW5NU7ZBM3VFN7MWMRB","download_json":"https://pith.science/pith/KTHALUMTW5NU7ZBM3VFN7MWMRB.json","view_paper":"https://pith.science/paper/KTHALUMT","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2112.08352&json=true","fetch_graph":"https://pith.science/api/pith-number/KTHALUMTW5NU7ZBM3VFN7MWMRB/graph.json","fetch_events":"https://pith.science/api/pith-number/KTHALUMTW5NU7ZBM3VFN7MWMRB/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/KTHALUMTW5NU7ZBM3VFN7MWMRB/action/timestamp_anchor","attest_storage":"https://pith.science/pith/KTHALUMTW5NU7ZBM3VFN7MWMRB/action/storage_attestation","attest_author":"https://pith.science/pith/KTHALUMTW5NU7ZBM3VFN7MWMRB/action/author_attestation","sign_citation":"https://pith.science/pith/KTHALUMTW5NU7ZBM3VFN7MWMRB/action/citation_signature","submit_replication":"https://pith.science/pith/KTHALUMTW5NU7ZBM3VFN7MWMRB/action/replication_record"}},"created_at":"2026-07-05T04:20:31.742576+00:00","updated_at":"2026-07-05T04:20:31.742576+00:00"}